TY - JOUR
T1 - Massively parallel nonparametric regression, with an application to developmental brain mapping
AU - Reiss, Philip T.
AU - Huang, Lei
AU - Chen, Yin Hsiu
AU - Huo, Lan
AU - Tarpey, Thaddeus
AU - Mennes, Maarten
N1 - Funding Information: The authors thank Eva Petkova, Ciprian Crainiceanu, Davide Imperati, Michael Milham, Clare Kelly, Babak Ardekani, and Xavier Castellanos for very helpful discussions; and the Editor, Associate Editor, and referees for valuable comments on the initial manuscript. The first author’s research is supported in part by National Science Foundation grant DMS-0907017 and National Institutes of Health grant 1R01MH095836-01A1.
PY - 2014
Y1 - 2014
N2 - A penalized approach is proposed for performing large numbers of parallel nonparametric analyses of either of two types: Restricted likelihood ratio tests of a parametric regression model versus a general smooth alternative, and nonparametric regression. Compared with naïvely performing each analysis in turn, our techniques reduce computation time dramatically. Viewing the large collection of scatterplot smooths produced by our methods as functional data, we develop a clustering approach to summarize and visualize these results.Our approach is applicable to ultra-high-dimensional data, particularly data acquired by neuroimaging; we illustrate it with an analysis of developmental trajectories of functional connectivity at each of approximately 70,000 brain locations. Supplementary materials, including an appendix and an R package, are available online.
AB - A penalized approach is proposed for performing large numbers of parallel nonparametric analyses of either of two types: Restricted likelihood ratio tests of a parametric regression model versus a general smooth alternative, and nonparametric regression. Compared with naïvely performing each analysis in turn, our techniques reduce computation time dramatically. Viewing the large collection of scatterplot smooths produced by our methods as functional data, we develop a clustering approach to summarize and visualize these results.Our approach is applicable to ultra-high-dimensional data, particularly data acquired by neuroimaging; we illustrate it with an analysis of developmental trajectories of functional connectivity at each of approximately 70,000 brain locations. Supplementary materials, including an appendix and an R package, are available online.
KW - Functional data clustering
KW - Neuroimaging
KW - Penalized splines
KW - Restricted likelihood ratio test
KW - Smoothing parameter selection
UR - http://www.scopus.com/inward/record.url?scp=84901816548&partnerID=8YFLogxK
U2 - https://doi.org/10.1080/10618600.2012.733549
DO - https://doi.org/10.1080/10618600.2012.733549
M3 - Article
SN - 1061-8600
VL - 23
SP - 232
EP - 248
JO - Journal of Computational and Graphical Statistics
JF - Journal of Computational and Graphical Statistics
IS - 1
ER -